CatUMap
JL
2024-01-03
df <- fread("../output/50.2.umap.tsv.gz")
df$latScale=squish(scale(df$lat),-2,2)
df$lonScale = squish(scale(df$lon),-2,2)
df$SpeedScale = squish(scale(df$Speed),0,50)
# map the lat lon to US states
library(maps)
library(mapdata)
metricTypes <- c("euclidean", "haversine")
# https://umap-learn.readthedocs.io/en/latest/embedding_space.html
# transformations to 2d
# x = np.sin(sphere_mapper.embedding_[:, 0]) * np.cos(sphere_mapper.embedding_[:, 1])
# y = np.sin(sphere_mapper.embedding_[:, 0]) * np.sin(sphere_mapper.embedding_[:, 1])
# z = np.cos(sphere_mapper.embedding_[:, 0])
#
# x = np.arctan2(x, y)
# y = -np.arccos(z)summary
skim(df)| Name | df |
| Number of rows | 148044 |
| Number of columns | 21 |
| Key | NULL |
| _______________________ | |
| Column type frequency: | |
| character | 5 |
| numeric | 15 |
| POSIXct | 1 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| Activity | 0 | 1 | 4 | 10 | 0 | 6 | 0 |
| Name | 0 | 1 | 2 | 3 | 0 | 2 | 0 |
| Notes | 0 | 1 | 0 | 8 | 137357 | 11 | 0 |
| UUID | 0 | 1 | 16 | 36 | 0 | 57 | 0 |
| Version | 0 | 1 | 0 | 28 | 19245 | 18 | 0 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| Accuracy | 0 | 1 | 2.045000e+01 | 16.22 | 2.68 | 8.00 | 15.23 | 3.20000e+01 | 9.991000e+01 | ▇▃▁▁▁ |
| Elevation | 0 | 1 | 2.618800e+02 | 292.58 | -450.00 | 146.36 | 255.12 | 2.73670e+02 | 1.133136e+04 | ▇▁▁▁▁ |
| Heading | 0 | 1 | 1.416500e+02 | 117.07 | -1.00 | 23.09 | 136.66 | 2.38750e+02 | 3.600000e+02 | ▇▃▅▃▃ |
| Pressure | 0 | 1 | 1.133100e+02 | 120.26 | 0.00 | 97.73 | 98.67 | 9.96600e+01 | 1.027110e+03 | ▇▁▁▁▁ |
| Speed | 0 | 1 | 1.460000e+00 | 7.22 | -1.00 | 0.00 | 0.00 | 0.00000e+00 | 2.881900e+02 | ▇▁▁▁▁ |
| UnixTime | 0 | 1 | 1.606792e+09 | 50904817.47 | 1525624972.00 | 1561565840.50 | 1604263022.50 | 1.65024e+09 | 1.703182e+09 | ▇▆▇▆▅ |
| lat | 0 | 1 | 0.000000e+00 | 1.00 | -8.83 | -0.92 | 0.55 | 5.50000e-01 | 3.950000e+00 | ▁▁▁▇▁ |
| lon | 0 | 1 | 0.000000e+00 | 1.00 | -5.93 | -0.11 | -0.03 | 1.60000e-01 | 1.099000e+01 | ▁▇▁▁▁ |
| umap_euclidean0 | 0 | 1 | 3.700000e-01 | 4.48 | -12.77 | -2.20 | 0.03 | 2.66000e+00 | 1.381000e+01 | ▁▃▇▂▁ |
| umap_euclidean1 | 0 | 1 | 5.610000e+00 | 4.43 | -7.70 | 3.14 | 5.76 | 8.20000e+00 | 1.900000e+01 | ▁▃▇▃▁ |
| umap_haversine0 | 0 | 1 | 1.100000e-01 | 18.64 | -135.22 | -9.08 | 0.19 | 9.54000e+00 | 1.322700e+02 | ▁▁▇▁▁ |
| umap_haversine1 | 0 | 1 | 5.690000e+00 | 11.67 | -71.32 | -0.16 | 5.73 | 1.16000e+01 | 8.908000e+01 | ▁▁▇▁▁ |
| latScale | 0 | 1 | 3.000000e-02 | 0.89 | -2.00 | -0.92 | 0.55 | 5.50000e-01 | 3.950000e+00 | ▅▁▇▁▁ |
| lonScale | 0 | 1 | 2.000000e-02 | 0.94 | -2.00 | -0.11 | -0.03 | 1.60000e-01 | 1.099000e+01 | ▇▁▁▁▁ |
| SpeedScale | 0 | 1 | 1.900000e-01 | 0.96 | 0.00 | 0.00 | 0.00 | 0.00000e+00 | 3.970000e+01 | ▇▁▁▁▁ |
Variable type: POSIXct
| skim_variable | n_missing | complete_rate | min | max | median | n_unique |
|---|---|---|---|---|---|---|
| Time | 0 | 1 | 2018-05-06 16:42:52 | 2023-12-21 18:02:03 | 2020-11-01 20:37:03 | 148042 |
Cat umaps
